StockWatch Focus, March 15, 1997
Cisco represents the quintessential high-tech investment: the stock has returned about 5,000% in just under seven years, but few non-technical stockholders or major fund managers could tell you about Cisco products in any level of detail. Public understanding of the bits, bytes, and flags involved in IP routing is immaterial, as Cisco has built itself into the de facto network infrastructure monopoly, creating what is close to a monopoly in the space. Its products range from routers and switches to remote access products to telecommunications equipment brought to the table by last year’s Stratacom acquisition.
Cisco gets more than its fair share of press, mostly generating statements about its aggressive acquisitions. In the past three years, Cisco has picked up 14 companies to round out its product line, using the purchasing power of a strong stock and a solid cash flow to fund the buying spree. CEO John Chambers graced the cover of the March 1997 Red Herring (see “Resources), accompanying an article describing Cisco’s success in glowing terms.
Is this Cisco kid a friend of mine? I think this is the time to look at Cisco again, from a software perspective and as a Java play. Confused yet? Read on. For the price sensitive, here’s the Cisco story in a nutshell. The stock has slipped about one-third since the beginning of the year in response to three salient events: First, Cisco missed its earnings number for the last quarter; second, 3Com’s announced purchase of US Robotics sank all of the networking players; and third, Hambrecht & Quist shot Cisco, Ascend, and Cascade just last week, dropping their ratings and expressing worries about future earnings. The terrible trio of market forces put CSCO right around 0 per share.
The downsides to the stock are well-described by the analysts: They doubt that Cisco can maintain its lofty gross margins (about 65 percent in the last quarter); industry movement toward IP switching instead of routing might erode Cisco’s strongest and most margin-rich product line; and the stock is trading at 47 times trailing earnings with a valuation of nine times sales. That’s a pretty rich price to pay for any stock, least of all one that few people understand. Microsoft sells for 11 times sales with a trailing P/E ratio of well over 30 — so it’s tempting to lump the monopolies into the same pricing model and rationalize nearly any price for Cisco.
Hopefully you want more than hand-waving. Maybe you even want to know how Java fits into the picture. Let me offer you five reasons to find Cisco attractive:
Cisco’s products increasingly are made from commodity components (Ethernet interfaces, memory, power supplies) coupled with Cisco software. Cisco’s IOS operating system represents value-add that is purely software, not subject to the margin and manufacturing pressures of a mostly hardware company.
Cisco has been perceived as being weak in the unifying management infrastructure arena, compared to other players that sell network management tools with their routers and switches. Fortunately, Java changes the landscape a bit just in time for Cisco — the market is looking more for browser-based, pluggable (some might say “open”) frameworks that use a variety of management protocols. Sun’s Java Management APIs (JMAPI) offer Cisco a chance to not just reach parity in the tools arena, but to set the bar higher in interoperability with existing tools and next-generation interfaces.
Address space conservation in the 32-bit, IP version 4 standard used today has been pushed off a year or so at a time, but the indisputable truth is that IP version 6 is coming and is bringing a whole host of new routing, switching, and address translation problems with it. I can summarize the effect in two words: demand creation. IP version 6 has features for class of service and traffic prioritization, both of which require management infrastructure and software (see points 1 and 2 above).
Cisco uses the Web well. Nearly three-quarters of its customer service is provided over the Internet, and Cisco has a rapidly growing ‘Net-based sales figure. Some estimates put Cisco’s sales over the Internet in the billion-dollar-plus range within a few years. Selling products through the ultimate of indirect channels means the cost of sales and administration goes down. Face it, a Web server costs about one-quarter of what a good sales person does and doesn’t take the afternoon off to play golf or bury good cigars and wine in expense reports! While Cisco’s gross margins (the price received for the product minus the cost of producing it) may be under pressure, Cisco can counter by reducing the selling and marketing costs, relieving the downward push on ‘Net income. This may be the first example of a sales model that seriously questions the utility of gross margins as an indicator of pricing sensitivity. Java is likely to play a larger role in Cisco’s online sales, just as it did for early-adopter National Semiconductor with its online product catalogs. Java applets that let you combine product browsing with capacity planning, network design, and auditing capabilities would make Cisco’s Web site truly a one-stop, self-service network supermarket.
- Java, Java-based network computers, and the emergence of push technologies like Marimba’s Castanet (see Resources) have made a number of companies evaluate their current and future network architectures. As popular as IP switching will be, it can’t overcome the politics of multiple departments wanting physical network isolation with routers, or eliminate internal security concerns that routers and firewalls address with ease. George Gilder often says that bandwidth is infinite, which would imply a zero price for it; fortunately for Cisco and MCI and friends, bandwidth is cheap but still costs money. Our demand for it, however, may be increasing at an alarming rate. Adoption of Java-based clients, and careful analysis of intranet/Web usage is driving companies to grow their network infrastructure.
Given the current market’s volatility and instant distaste for anything remotely technical, I’m going to hold off on adding Cisco to our portfolio until it announces its next quarter’s earnings. If Cisco looks like it’s back on the financial track, we’ll make it a friend indeed.
